Time series are data observed over time (either in continuous time or at discrete time periods).

learn more… | top users | synonyms

1
vote
0answers
24 views

Seasonal ARIMA for each weekday instead of double seasonal ARIMA

I have read some papers on forecasting time series with double seasonality (e. g. hourly data with daily and weekly seasonality). I understand that double seasonal ARIMA can be used for that purpose. ...
0
votes
1answer
20 views

Running Linear Regression Model

my question is can i run linear regression model using summarized counts/frequencies? For example, my dependent variable is total number of people who is aware of a specific TV show and my independent ...
0
votes
0answers
19 views

Detect increase/decrease events on time series

Given a time series, I have to detect two types of events: 1) "medium" decrease 2) "high" increase Event detection should be "fast enough". I used quotation marks as I'd like to set different ...
0
votes
0answers
26 views

Normalizing year data

I have a data set in which there are data from April 2010 to Dec 2010 ( 9 Months) Jan 2011 to Dec 2011 (12 Months ) Jan 2012 to Dec 2012 (12 Months) Jan 2013 to Dec 2013 (12 Months) Jan 2014 to ...
1
vote
0answers
27 views

Contradiction in the ADF (Augmented Dickey-Fuller) and KPSS (Kwiatkowski–Phillips–Schmidt–Shin) tests for financial time series

I use the ADF and KPSS to test for stationarity / non-stationarity of price increments in financial time series. The two test applied provide different results for low lags, but the same result for ...
1
vote
1answer
19 views

Predicting time series value given a threshold weight

I have 2 datasets. One is time series data of sale of homes by region by type: ...
0
votes
0answers
34 views

Identify outlier usage intervals in time-series data

I want to find outliers in power consumption in real-time, at hourly rate, i.e., at the end of the hour, I should say whether power consumption in current hour was outlier/anomalous or not. Approach: ...
0
votes
1answer
29 views

VAR lag length vs Johansen cointegration test outcome?

First puzzle: I am taught that the lag order of VECM does not affect the cointegration rank because the lag order is for the differenced regressors. But, I see the contrary: I experimented with sample ...
1
vote
0answers
19 views

Prewhitening Regressors in Lagged Time Series Regression

I'm trying to identify significant lags in a time series regression such that $Y = \beta_0X_t + \beta_1X_{t-1} + ... + \beta_iX_{t-i} + \alpha_0Z_t + \alpha_1Z_{t-1} + ... + \alpha_jZ_{t-j}$ I ...
1
vote
0answers
44 views

Maximum Likelihood estimation and the Kalman filter

I know the Kalman filter recursions and can derive these but what I don't really get is how to estimate the hyper parameters using maximum likelihood. I understand that when running the Kalman filter ...
0
votes
0answers
14 views

Testing a non-AR time series for mean reversion

If one discretizes Heston's dynamics for the instantaneous variance of the stock price, one gets the following time series : $$ V_k - V_{k-1} = a + b V_{k-1} + c \sqrt{V_{k-1}} \mathcal{N} \left( 0, ...
0
votes
0answers
22 views

fitting a dynamic bayesian model to irregular time data

I have a dynamic epidemiological model which I solve with scipy's ODEint and fit to my data using pymc. My data is irregular in ...
1
vote
0answers
30 views

How many observations to estimate a parameter of an Archimedean copula?

Let's consider for example the bivariate Gumbel copula. $$C(u_1, u_2)=exp \left[-\left((-ln(u_1))^{\theta}+(-ln(u_2))^{\theta}\right)^{\frac{1}{\theta}}\right]$$ In R there are some functions (such as ...
0
votes
0answers
8 views

Linear regression (best fit line) on moving averages vs raw data?

I have a series of sets of data over a period of time, with the amount of data available being quite variable between sets. One has points for almost every day but is really quite noisy; another has ...
0
votes
1answer
20 views

Hourly electricity demand data finding AR and MA terms [duplicate]

I am new to time series analysis. I have hourly electricity demand data for five years (having multiple seasonalities at intra-daily, intra-weekly, and annual periodicities) and I want to guess the ...
0
votes
0answers
28 views

Measuring the effect of weather on retail sales

I'm currently working on modeling this as an ad hoc. Sr mgmt want to know how much of our sales growth during the year can be attributed to weather. I chose to investigate "weather" as temp & ...
1
vote
0answers
10 views

Can I have a time covariate to account for temporal trends as well as time series biases such as AR/MA?

Can I have a time covariate to account for temporal trends as well as time series biases such as AR/MA? For instance if I have a model: Y ~ X + t where t is time, then can this be an alternative ...
0
votes
1answer
27 views

Time as random effect or fixed effect in glmmADMB

I have a longitudinal dataset where patients have a measurement with a date, currently coded as time from end of treatment (days). Now, I want to build a model. Roughly, a zero inflated Poisson model ...
0
votes
0answers
11 views

Should we predict residual of ensemble model and add it to final prediction?

We have been doing a time series project on daily wise data. We built 5 different models SVM, XGBoot, ARIMA, KNN, ANN. We then built predictive models for residuals for all of these models and got ...
0
votes
1answer
25 views

Time Series Forecasting, Log or non-Log

I have read that you should use log transformations when the fluctuations on your data are increasing over time, but what do you do if the fluctuations level out over time? A plot of the time series(...
0
votes
1answer
20 views

Time Series Modelling[Issue with modelling the residuals]

I am doing the sales forecast. I found the trend and seasonality manually for my time series data. Regressed time series data against the trend and seasonality and found the residuals. The residuals ...
0
votes
1answer
35 views

Interpreting Time series regression/bivariate sorts

I am somewhat unsure how to interpret some result from an analysis that I have done on two independent variables and a dependent variable. My goal is to test whether the abnormal return difference on ...
0
votes
1answer
10 views

Imputation for Time Series of Accumulated Value

I have a regular time series of accumulated values of a variable (usage) with some missing (sometimes consecutive) intervals. Is there an imputation method that methodologically considers this ...
1
vote
0answers
10 views

DCC models in R: how is the first starting value chosen?

The DCC model is defined through the proxy $Q_t$ as $$Q_t=(1-\alpha-\beta) \overline{Q} +\alpha\epsilon_{t-1}\epsilon_{t-1}' + \beta Q_{t-1}$$which is then normalized to find the correlation matrix $...
1
vote
0answers
8 views

Fractional Gaussian noise, the KPSS test, and stationarity

Fractional Gaussian noise (fGn) is characterized by the mean ($\mu$), the standard deviation ($\sigma$), and the Hurst index ($H$). It's my understanding that it is stationary, for the simple reason ...
0
votes
1answer
35 views

How to determine seasonality of a binary variable?

I have a dependent binary variable Y, and an independent date variable X. I want to find out if there is any seasonality (at the year level). A few notes: The binary variable is in my model non-...
1
vote
1answer
28 views

How does trend stationary recovers from shocks in long run?

I was trying to understand difference between drift and trend wherein I came across concepts of unit roots and trend stationary. (I haven't read any books on time series, just going through web). ...
0
votes
0answers
20 views

Threshold cointegration

I have a panel data N=45 T=25. Engle-Granger test confirms co-integration between two I(1) variables. I would like to test for threshold cointegration between these two vars. Is there a user written ...
3
votes
0answers
44 views

Decomposition of SARIMA models

I use R for time series analysis. I would like to evaluate decomposition algorithms. decompose and stl from "stats" package lead ...
4
votes
1answer
107 views

Why is my kalman filter trusting so much my observations?

This question follows the one asked there. I am trying to filter an equity index (Stoxx 600) time series using kalman filter. I'm using the R package dlm and my code is inspired from the dlm ...
1
vote
0answers
35 views

Best step pattern for unconstrained/subsequence DTW(Dynamic Time Warping)

I am implementing an unconstrained/subsequence DTW algorithm (in R). The query of data which I am trying to find a match within the reference data is much smaller (as compared to reference). I have ...
0
votes
0answers
13 views

Critical point in time series and pattern comparison

I have some data from a recent experiment. Two factors, each had 2 levels, resulting in 4 conditions. In each condition there were 12 participants, so 48 participants in total. Each participant ...
4
votes
1answer
45 views

Time series analysis of electricity load questions

I have hourly data of electricity load (MW) that span 8 months (that is, 5760 data points). I also have predictions from a regression model for the same period. My goal is: to examine some ...
1
vote
1answer
44 views

When to run truncated backpropagation through time in recurrent neural networks?

I'm interested in training recurrent neural networks using truncated packpropagation through time (BPTT). From Sutskever (2013): Truncated BPTT...processes the sequence one timestep at a time, ...
2
votes
0answers
40 views

What is the difference between the Monte Carlo Method in R package 'DMwR' and a normal Monte Carlo Method?

I am trying to estimate the performance of a machine learning model on time series data. I saw the example of model evaluation using Monte Carlo Estimates from the book "Data Mining With R Learning ...
0
votes
1answer
22 views

Generalized Hurst Exponent - What value to specify for $\tau_{\max}$?

Consider a time series $X: S \to \mathbb{R}$, where $S := \{\nu, 2\nu, 3\nu, \ldots T\}$, and $T$ is a multiple of $\nu > 0$. For each $\tau \in (0, \tau_{\max}] \cap S$ and $q \in \mathbb{N}$, ...
0
votes
2answers
21 views

Scaling predictors in ARIMA model

If a predictor in an ARIMA model has much lower magnitude than the variable you are trying to predict, then do you need to multiply it by a scalar in order for it to be an effective predictor in the <...
0
votes
0answers
33 views

Why do VAR forecasted values radically change depending on which month historical data end?

I am building a model to forecast housing variables using vector autoregression. I am encountering spurious results. My forecasted values change dramatically depending in which month the historical ...
0
votes
0answers
14 views

Negatively Correlated Predictors in Arima Model

If a predictor is negatively correlated with a variable you are trying to forecast in an Arima model, will Arima pick up the negative correlation when you add the predictor in the xreg argument? Is ...
0
votes
0answers
7 views

Combining Lower Correlation Predictors to Create Higher Correlation Predictors

I'm working on an Arima model to forecast a given variable and so I'm looking in my data for variables with correlation to the variable I'm trying to predict, to add as predictors in the xreg argument....
0
votes
1answer
23 views

Does picking randomly from an imbalanced group balance the group?

If I had a bag of marbles with 75% blue and 25% red (ratio is what matters not raw number, so this applies to 100 marbles, 1000 marbles, 100000 marbles) So if I had this imbalance of 75% blue and 25% ...
0
votes
1answer
70 views

Difference between SUTSE (Seemingly Unrelated Time Series Equations) and SUR (Seemingly Unrelated Regressions)

I am studying time-series econometrics and in particular Dynamic Linear Models for multivariate time-series. Someone can help me in understanding which is the difference between SUTSE (Seemingly ...
1
vote
2answers
65 views

Lag-free filter methods for time series

I'm currently working with accelerometer based raw data (100 hz). Now I want to low pass filter this timeseries of accelerations for further analyses. I tried different filters like the simple moving ...
1
vote
1answer
37 views

General-to-specific subset selection (“Autometrics”) performing well in macroeconomics

I wonder why general-to-specific (GETS) subset selection and particularly the Autometrics algorithm are performing well in macroeconomic modelling/forecasting. How does Autometrics work? Doornik "...
0
votes
0answers
22 views

How we can find trend of time series and turning points?

I want something like this in a time series: Currently I'm using some linear interpolation to find trend and turning points. What other methods can I use to find these turning points in a time ...
0
votes
0answers
24 views

What is Maximal Lyapunov exponent and how to calculate and interpret it in R?

I was researching about chaotic time series and came across Maximal Lyapunov exponent. I tried to read many articles regarding it but the only thing I understood about MLE is this is used to check ...
0
votes
1answer
42 views

Regression result for second model still shows insignificant variables

I run a regression for my study. I used a quarterly data included 33 number of observations. my problem is independent variables are significant but not perfectly significant. when i run a ...
0
votes
0answers
9 views

Time-varying predictive model for a set of proportions

Suppose there is a casino where people bet on a weekly horse race. On Sunday, the casino publishes the prices for a wager on each horse for the upcoming Saturday's race. Everyone who wagers on the ...
0
votes
0answers
22 views

What is the correct measure of similarity for two supposedly identical time series?

I have two time series of prices (actually, bid quotes) from different sources, and I want to know how similar they are. I've calculated simple Pearson product-moment correlation between the prices, ...
2
votes
0answers
20 views

leave-h-out cross validation

I'm doing multistep forecasts of univariate time series and a wide range of exogenous leading indicator variables are available. Therefore I'm looking for ways to optimally select and/or combine ...